* [AutoParallel] quantization pass support export

* support subgraph

* move_presist_var_to_global_block

* update unittest

* fix ci-coverage

* fix codestyle

* fix fake_dequantize_op

* remove unused var

* fix ci error and aprroval error

* add unittest for fp16 in test_dequant_linear

* replace mutable data

* fix unittest in non-cuda-core

* fix unittest
Co-authored-by: Ncarryyu <569782149@qq.com>
Co-authored-by: Nwufeisheng <wfs1997@163.com>

* delete old dygraph sharding

* cinn use interpretercore

* fix bug

* fix compile bug

* fix scope bug

* refine code

* refine code by comment

* refine code by comment

* Support priority scheduling for standalone executor

* Add CPU test

* rm in_legacy_dygraph nn part1

* rm non_static_mode

* modify rrelu

* delete _in_legacy_dygraph

* feat: broadcast_object_list & scatter_object_list

* chore: update ut conf

* get_backend & is_available

* docs: update requirements

* fix: resolve conflicts
Co-authored-by: NLiYuRio <liyuruijx@163.com>

* rm legacy unittest part5

* add custom op

* migrate shaple sgd, split,sign xpu kernels to phi, test=kunlun

* fix dlrm throughput problem, test=kunlun

* add index sample fp16 support

* remove fluid APIs in distributed_strategy.py and role_maker.py

* Revert "remove fluid APIs in distributed_strategy.py and role_maker.py"

This reverts commit 223bbee990d3bf69e252fc3c0f19e3873550a264.

* move load_inference_model to distributed

* fix origin develop codes diff

* move _endpoints_replacement

* delete line

* reset line

* add unittest case of load_inference_model

* fix unittest

* fix unittest

* fix coverage

* fix coverage

* revert concat and change concat to stack

* let stack kernel support int8, uint8 and bool type

* [Auto Parallel] Rename methods of ProcessMesh

* [Auto Parallel] Impl the python process_mesh by the c++ one

* [Auto Parallel] Add some minor modifications

* [Auto Parallel] Rename some methods

* [Auto Parallel] Remove unnecessary codes

* [Auto Parallel] Add back some removed files

* [Auto Parallel] Fix bugs

* [Auto Parallel] Fix a bug

* Update process_mesh.cc

* [Auto Parallel] Fix a bug

* remove loss.py

* fix __init__.py

* Memory search for stream analyzer

* Shrink redundant waiters

* Enable identity_scale_op_clean_pass by default

* IdentityScaleOpCleanPass assert_var_not_persistable

* delete legacy dygraph code in python/paddle/distributed

* refine

* support cinn ci

* modify .whl dir

* support CINN

* support cinn

* support kunlun200 and kunlun200test

* support cinn

* fix tests in kunlun200

* fix bug that can't find libxpu.so.1

* test

* 修改linspace参数"stop"的介绍

* 更新了conv2d_transpose英文文档中 1.公式异常 2.参数说明异常 3.padding=SAME和VALID的公式说明; test=docs_preview

* 解决了
1.英文版Note中的api_fluid_layers_conv2d的链接失效，
2. 英文版参数中的padding_start的参数错误
3. 参数中bias_attr和param_attr的链接失效
4. 参数缺少optional

* 更新了paddle.static.auc的英文文档的Return描述以及函数的参数

* Update python/paddle/tensor/creation.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* 更新了shard_index的ignore_value描述

* 修改了英文文档的paddle.static.nn.conv3d_transpose的公式介绍

* add py_func COPY-FROM label; test=document_fix

* Update python/paddle/tensor/manipulation.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* formula; test=document_fix

* formula; test=document_fix

* formula; test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* memcpy overlap

* memcpy newexe

* fix docstrings:
1. 修复了 python/paddle/nn/layer/conv.py 中出现的 Tex 语法问题；
2. 为 python/paddle/nn/functional/conv.py 修复 Tex 语法问题，修改 note 及参数说明格式，尝试添加论文链接；
3. 修改 python/paddle/nn/layer/loss.py、python/paddle/nn/functional/loss.py 中参数说明的格式；
4. 修改 python/paddle/nn/layer/common.py、python/paddle/nn/functional/common.py 中参数说明的格式，添加 optional 说明。

* fix docstring format, add paper's hyperlink.

* fix formula Tex.

* fix format error; test=docs_preview

* fix reference hyperlink; test=docs_preview

* fix docstring; test=docs_preview

* fix Tex format.

* fix api reference;  test=docs_preview

* test=document_fix

* Update conv.py

* formula; test=document_fix

* test=document_fix

* formula; test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* fix matmul double and triple grad

* remove some comment

* add matmul_double_grad unit test

* fix matmul triple grad

* fix dot triple grad and add unit test

* modify codestyle

* fix dot_grad

* refactor dot triple grad

* disable some unit test

* fix unit test

* fix unit test in double grad

* rm paddle.fluid.contrib.layers.BasicLSTMUnit basic_lstm BasicGRUUnit basic_gru

* rm dependency in __init__.py

* rm eager guard test

* retain grad for xpu test

* fix custom double_grad backward=None

* fix custom_relu.cu bug && polish testcase of double_grad

* remove old dynamic graph test

* add warp transducer code

* add ouutput(CLSInds) for fused_token_prune